FathomVerse: A community science dataset for ocean animal discovery

TL;DR

FathomVerse v0 is a novel deep-sea animal detection dataset with challenging images and categories, designed to advance computer vision research in ocean exploration and ecological understanding.

Contribution

The paper introduces the first deep-sea animal detection dataset with 3843 images and 8092 bounding boxes across 12 categories, addressing a new vision challenge in marine biology.

Findings

Dataset enables research on fine-grained transfer learning.

Supports studies on species distribution and ecological modeling.

Challenges existing computer vision models with complex deep-sea imagery.

Abstract

Can computer vision help us explore the ocean? The ultimate challenge for computer vision is to recognize any visual phenomena, more than only the objects and animals humans encounter in their terrestrial lives. Previous datasets have explored everyday objects and fine-grained categories humans see frequently. We present the FathomVerse v0 detection dataset to push the limits of our field by exploring animals that rarely come in contact with people in the deep sea. These animals present a novel vision challenge. The FathomVerse v0 dataset consists of 3843 images with 8092 bounding boxes from 12 distinct morphological groups recorded at two locations on the deep seafloor that are new to computer vision. It features visually perplexing scenarios such as an octopus intertwined with a sea star, and confounding categories like vampire squids and sea spiders. This dataset can push forward research on topics like fine-grained transfer learning, novel category discovery, species distribution modeling, and carbon cycle analysis, all of which are important to the care and husbandry of our planet.